* Mike Gerdts <[EMAIL PROTECTED]> [2008-01-15 05:20]:
> On Jan 14, 2008 12:09 PM, Stephen Hahn <[EMAIL PROTECTED]> wrote:
> Unless there is strong objection, I'll only change dupnode to use a
> list and not change the rest.
Sounds good.
> > 3119 - 3124. How does this comment match up with the stat()? (The
> > other question it raises is that of a directory tree that doesn't
> > deliver an explicit bin, but has manual pages in the correct
> > location: the if (stat()) test eliminates handling such a tree.)
> >
> > 3127. (Nit.) I thought 3120 was "first"? :)
>
> Collapsed to one "first" comment. The intent of deriving MANPATH from
> PATH is (mostly) to give the section 1* pages that are most likely to
> correspond to the command that is executed. If bindir doesn't exist,
> it doesn't seem to make sense to include the MANPATH at that point.
> For those people that have a need to search man directories that don't
> have corresponding bin directories, the following is recommended.
>
> PATH=...
> export PATH
> unset MANPATH
> MANPATH=`man -p`:/my/extra/man
> export MANPATH
>
> Would you find it better to allow people to trick man by setting PATH
> to point to non-existent bin directories to pick up those man
> directories?
>
> That is:
>
> if (stat(bindir, &sb) == 0) {
> for (i = 0 ...
> ...
> }
No, your recommendation is fine. I was just working through what the
code does and doesn't do.
Cheers
Stephen
--
[EMAIL PROTECTED] http://blogs.sun.com/sch/
_______________________________________________
opensolaris-code mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code
